EU and UK rules tighten oversight of Artificial Intelligence hiring tools

US employers using Artificial Intelligence in recruitment across Europe face stricter oversight under the EU Artificial Intelligence Act, GDPR, and the UK’s Data (Use and Access) Act 2025. Hiring tools that score, rank, or screen candidates are drawing closer scrutiny for bias, transparency, and meaningful human review.

Regulators in the European Union and the United Kingdom are increasing scrutiny of algorithmic discrimination in employment decisions by combining new Artificial Intelligence specific rules with existing data protection and anti-discrimination laws. For US employers, that creates a more demanding compliance environment for recruitment, candidate screening, promotion, performance evaluation, and some worker monitoring tools. Artificial Intelligence systems used in these areas can no longer be treated as simple vendor products because employers are expected to understand how the systems were trained, how fairness is tested, and which legal constraints apply in each jurisdiction.

Under the EU Artificial Intelligence Act, most tools used in employment and worker management are classified as high-risk because they can directly affect workers’ livelihoods. High-risk systems must meet obligations before deployment, including documented risk management, data governance and quality controls, technical documentation, logging, transparency, and meaningful human oversight. Article 10 of the AI Act focuses on data quality and bias. High-risk HR Artificial Intelligence must be trained, validated, and tested on data that are relevant, representative, sufficiently diverse, and as free of errors as possible. Organizations are also expected to carry out systematic bias testing with documented mitigation and ongoing monitoring.

In the EU, those rules operate alongside the GDPR. Article 22 restricts decisions based solely on automated processing that produce legal or similarly significant effects, such as hiring and termination. Employers that rely heavily on automated scoring or ranking must therefore ensure meaningful human involvement, provide candidates with information about the logic involved, and offer routes to contest decisions. The European Court of Justice’s SCHUFA decision from 2023 reinforced a broad reading of Article 22 by treating automated profiling as an automated decision where third parties rely heavily on the resulting score. In hiring, that raises the risk that Artificial Intelligence generated scores or rankings will be treated as automated decisions even when a human only nominally approves the outcome.

The UK has taken a different path. The Data (Use and Access) Act 2025 amends the UK GDPR and Data Protection Act 2018 rather than mirroring the EU Artificial Intelligence Act. DUAA focuses on significant decisions taken solely by automated means, with the strictest restrictions applying where those decisions are based wholly or partly on special category data and where safeguards are lacking. The law simplifies some compliance duties, including record-keeping and certain assessments for re-use of personal data, and introduces limited categories of recognized legitimate interests. Even so, employers still need risk assessments, impact assessments, and safeguards around significant automated decisions. The ICO is also sharpening expectations around bias testing, transparency, explainability, complaint handling, and human involvement, while the Equality Act 2010 continues to prohibit direct and indirect discrimination.

Employers operating across the EU and UK should inventory their HR related Artificial Intelligence tools, determine which systems are high-risk or subject to automated decision rules, and require vendors to provide bias and data quality documentation. They should also conduct pre-deployment and periodic testing, document remediation where disparities appear, and design workflows in which humans genuinely review and can override Artificial Intelligence outputs. In several European countries, introducing Artificial Intelligence based HR tools may also require early engagement with works councils or comparable employee bodies, alongside clear notices, explanation rights, and complaint channels for candidates and employees.

68

Impact Score

AMD Instinct MI355X passes 1M tokens/sec in MLPerf 6.0

AMD says its MLPerf Inference 6.0 submission combined competitive single-node performance, multinode scale, and broader partner reproducibility. The company also highlighted first-time workloads and a heterogeneous submission spanning different systems and geographies.

Mistral Artificial Intelligence secures 830 million for Paris data center

Mistral Artificial Intelligence has raised 830 million in debt financing to operate a new data center outside Paris as it pushes for a larger independent cloud and compute footprint in Europe. The facility will be powered by thousands of Nvidia chips and forms part of a broader regional expansion plan.

Anthropic signals cybersecurity push with Mythos

Anthropic’s reported Mythos model points to a deeper push into cybersecurity and a broader effort to expand beyond coding-focused offerings. Enterprises are likely to treat it as one option among many rather than a single answer to their Artificial Intelligence needs.

Nvidia targets laptops with Arm chips at Computex 2026

Nvidia is preparing to introduce Arm-based laptop processors at Computex 2026 as it pushes into the consumer PC market. The new chips, developed with MediaTek, are aimed at thin-and-light systems with strong graphics and Artificial Intelligence performance.

OpenAI reports lower hallucination rates for GPT-5

OpenAI says GPT-5 produces fewer false claims than earlier models, especially when it can browse the web. The gains look smaller without web access, underscoring how much reliability still depends on live sourcing.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.